neuralmonkey.evaluators.mse module¶
-
class
neuralmonkey.evaluators.mse.MeanSquaredErrorEvaluator(name: str = None) → None¶ Bases:
neuralmonkey.evaluators.evaluator.SequenceEvaluatorMean squared error evaluator.
Assumes equal vector length across the batch (see SequenceEvaluator.score_batch)
-
static
compare_scores(score2: float) → int¶ Compare scores using this evaluator.
The default implementation regards the bigger score as better.
Parameters: - score1 – The first score.
- score2 – The second score.
- Returns
- An int. When score1 is better, returns 1. When score2 is better, returns -1. When the scores are equal, returns 0.
-
score_token(hyp_elem: float, ref_elem: float) → float¶ Score a single hyp/ref pair of tokens.
The default implementation returns 1.0 if the tokens are equal, 0.0 otherwise.
Parameters: - hyp_token – A prediction token.
- ref_token – A golden token.
Returns: A score for the token hyp/ref pair.
-
static
-
class
neuralmonkey.evaluators.mse.PairwiseMeanSquaredErrorEvaluator(name: str = None) → None¶ Bases:
neuralmonkey.evaluators.evaluator.EvaluatorPairwise mean squared error evaluator.
For vectors of different dimension across the batch.
-
static
compare_scores(score2: float) → int¶ Compare scores using this evaluator.
The default implementation regards the bigger score as better.
Parameters: - score1 – The first score.
- score2 – The second score.
- Returns
- An int. When score1 is better, returns 1. When score2 is better, returns -1. When the scores are equal, returns 0.
-
score_instance(hypothesis: List[float], reference: List[float]) → float¶ Compute mean square error between two vectors.
-
static